RaBitQ 1-bit quantized vector index in WebAssembly — 32× embedding compression with high-recall rerank, for browsers, Cloudflare Workers, Deno, and Bun
Get value at specific positions of a vector
RAG vector index helpers for HNSW and IVFFlat
RAG vector index helpers for HNSW and IVFFlat
RaBitQ 1-bit quantized vector index in WebAssembly — 32× embedding compression with high-recall rerank, for browsers, Cloudflare Workers, Deno, and Bun
Stores document embeddings in an OpenSearch vector index.
Deterministic Vector Index Factory - MCP Server
An HTTP/REST based Vector DB client built on top of Upstash REST API.
Curated collection of data structures for the JavaScript/TypeScript.
Edge-computed semantic search engine — WASM embeddings, vector index, and hybrid search in the browser
Slice GeoJSON data into vector tiles efficiently
Deterministic Vector Index Factory - MCP Server
Slice GeoJSON data into vector tiles efficiently
AWS SDK for JavaScript S3vectors Client for Node.js, Browser and React Native
Ultra-low-latency C++23 memory kernel for Node.js — Atlas vector index + Trace episodic WAL
MongoDB Atlas Search Index JSON schemas
Built-in support for popular icon fonts and the tooling to create your own Icon components from your font and glyph map. This is a wrapper around react-native-vector-icons to make it compatible with Expo.
Find the nearest point to a sample point
Parses vector tiles
Lightweight browser-based semantic search library with HNSW vector index and transformer embeddings
A lightweight, file-backed vector database for Node.js and browsers with Pinecone-compatible filtering and hybrid BM25 search.
Astronomy calculation for Sun, Moon, and planets.
Isomorphic storage client for Supabase.
Serialize mapbox vector tiles to binary protobufs in javascript.
Generic HNSW vector index with pluggable distance metrics.
Sliced-Wasserstein (SW₁) distance over discrete distributions, with optional vector-index Metric impl.
CLI for opensession.io - discover, upload, and manage AI coding sessions
Abstraction for indexing and searching vectors
gromit uses Redis and OpenAI embeddings to index your documentation
vapey uses Redis and OpenAI embeddings to index your documentation
Interface to control vector databases settings (like indexes, collections, etc).
Ruby client library which includes index and vector operations to upload embeddings into Pinecone and do similarity searches on them.
A Ruby library for text classification featuring Naive Bayes, LSI (Latent Semantic Indexing), Logistic Regression, and k-Nearest Neighbors classifiers. Includes TF-IDF vectorization, streaming/incremental training, pluggable persistence backends, thread safety, and a native C extension for fast LSI operations.
Crowsad was conceptualized as a way of blending Duck Duck Go with Dangling Modifier generation so as to learn new words overs time, rather than having to index everything in one go. It uses four input vectors and one output vector. Github version is old version. Credit Andrew Jones for Duck Duck Go.
Ragnar is a high-performance RAG system that leverages Rust libraries through Ruby bindings for embeddings, vector search, and topic modeling. It provides a complete CLI for indexing documents and querying with LLMs.
MCP server that indexes codebases using AST-aware chunking and vector embeddings, providing semantic search for Claude Code and other MCP clients.
Add vector search to your Ruby apps without external services. zvec provides native bindings to Alibaba's high-performance C++ vector database via Rice, supporting HNSW, IVF, and flat indexes with multiple distance metrics. Build semantic search, recommendations, RAG pipelines, and similarity matching with pure Ruby — no HTTP APIs, no infrastructure, no latency overhead.
LEANN (Lightweight Embedding-Aware Neural Neighbor) is a Ruby gem for building and searching vector indexes with minimal storage. It provides semantic search and RAG capabilities with a beautiful, simple API. Supports multiple embedding providers: RubyLLM, OpenAI, Ollama, and FastEmbed.
Proper related posts plugin for Jekyll - uses document correlation matrix on TF-IDF (optionally with Latent Semantic Indexing). Each document is tokenized and stemmed, every word found is treated as keyword for analysis (except for some stop words). TF-IDF matrix for the whole site is calculated (including extra provided weights), then if given accuraccy is lower than 1.0, LSI algorithm is used to compute new simplified vector space. Document correlation matrix is created using dot product of the matrix and its transpose. For each of the post' related documents are inserted into priority queue (sorted by score from document correlation matrix), assuming the score is greater than minimal required score. Selected few bests related posts are retrieven from the queue. Liquid template for each post is rendered and <related-posts /> is replaced with the outcomes of algorithm.
No description provided.
No description provided.
No description provided.
No description provided.
No description provided.
No description provided.
No description provided.
No description provided.
No description provided.
No description provided.
No description provided.
No description provided.